ilastik
the interactive learning and segmentation toolkit
Leverage machine learning algorithms to easily segment, classify, track and count your cells or other experimental data. Most operations are interactive, even on large datasets: you just draw the labels and immediately see the result. No machine learning expertise required.
Train using simple brush strokes
Semantic segmentation using predicted class probabilities
Predict the semantic class of objects
Extract objects from a 3D volume semi-automatically
Select from a simple set of predefined image features
Count densely packed cells
Automatic and manual tracking of dividing cells (2D+3D)
Get in touch with other users at forum.image.sc or contact the friendly ilastik team directly.
We are happy to announce the release of version 2.0 of our Fiji plugin. The plugin allows you to conveniently import and export image in HDF5 format, as well as running trained Pixel Classification, Autocontext, Object Classification, Multicut, and Tracking Workflows from within Fiji.
Starting with beta version 1.4.1b5, we are distributing ilastik with native builds for apple silicon computers.
For beta version 1.4.1b1, we have added a new workflow that combines shallow and deep learning approaches.
Check out the documentation for the Trainable Domain Adaptation Workflow.